SECOM Intelligent Systems Laboratory
SECOM Japanese Site
About ISL Research Areas People Publications Careers Contact ISL

Ulas Bardak

The reason I applied to SECOM
 I have always been interested in the culture and traditions of Japan and even took a few related classes while I was working on my Ph.D. at Carnegie Mellon University. However, it was not until I met Dr. Hasegawa (my current group leader) at a conference in Taiwan that I started thinking seriously about working in Japan. After searching online for information on the quality and variety of research done at SECOM Intelligent Systems Laboratory (ISL), and learning about the size of SECOM's presence in Japan, Asia and Europe in general, I decided to apply. I was fortunate enough to be accepted and I have been working at the ISL since October 2007.

What has it been like working at SECOM?

 As my dissertation work was in improving optimization with applications in scheduling, I immediately joined the Scheduler Applications Group. My work in this group was focused on improving the functionality of SECOM's "eKakushin Scheduler", a scheduling application being used by many customers including hospitals for automatically generating employee rosters. This problem of producing rosters in response to complex conditions which may have contradicting goals is an active area of research. Initially, I explored different optimization approaches that could be used instead of the current optimization method, which gave me a chance to gain valuable knowledge about the system itself.picture

 After becoming familiar with the system, I took on a much bigger task ­ building a completely new, proof of concept version of the system that not only leverages new computer technologies but also tries to produce better schedules. I am currently working on this task which includes designing and implementing the relevant algorithms, the API between the different components of the system, and the components (such as the UI and the backend) themselves.

 Even though the project I am working on enables me to address the practical side of problem solving with actual implementation of software, this is not the only focus of my work. In fact, research and development of novel ideas is a central part of the work at the ISL. The approach I am currently developing for improving the results of optimization by automatically tuning the representation of constraints resulted in an academic paper in my first year and I am currently working on writing another paper with data from a new set of evaluations with different subjects. It is also very common (and in fact encouraged) for researchers at ISL to apply for patents which I am hoping to do as well.

 The fact that the ISL is not a fundamental research lab, but is geared towards research and development, means one can gain experience beyond his or her desk at the lab. For example, I got a chance to travel to and interview some of our target users in order to gain insight into their needs and thoughts about the behaviour of the current system. I have also built different versions of the current system, implementing an infrastructure for the features we wanted to evaluate and performed both on and off-site evaluations. I believe that channeling research into building a product that will be used by many users is a great experience and I am very happy to receive this experience at the ISL.


Has living in a completely different culture been challenging?
 Living and working in a foreign country can be a daunting task and I am glad SECOM has been there to make that task as easy as possible for me. Adjusting to Japan has definitely been an experience in itself but thanks to support staff at SECOM, my colleagues and my friends, it has been relatively smooth.
Before coming to Japan I was worried about so many things ­ "Will I be able to communicate using my primitive level of Japanese?" (Even though I took a few Japanese language classes at Carnegie Mellon during earning my Master's degree, I had all but forgotten my vocabulary), "Will I be able to follow the Japanese style of working?"... Once I came to Japan and started work all of these questions were answered. I was able to work with my colleagues in English which helped me adjust and be effective much quicker. SECOM has also been instrumental in arranging Japanese classes to improve my Japanese skills. I also did not have any problems with following the Japanese style of working which emphasizes group work and harmony ­ two concepts I have always followed in my work. All of these have helped me adapt to living and working in Japan.

This list includes publications which were published before entering SECOM.
· U. Bardak, S. Hasegawa, and T. Saito. PointFix: Learning from Fixing Individual Condition Violations. Submitted to IEEE CI-Sched 2009
· U. Bardak. Information Elicitation for Improving Optimization: Applications in Scheduling Problems, VDM Verlag. August 2008. ISBN: 3639079086.
· U. Bardak, E. Fink, C. R. Martens, and J. G. Carbonell. Scheduling with uncertain resources: Elicitation of additional data. IEEE International Conference on Systems, Man, and Cybernetics 2006.
· U. Bardak, E. Fink, and J. G. Carbonell. Scheduling with uncertain resources: Representation and utility function. IEEE International Conference on Systems, Man, and Cybernetics 2006. (Best Student Paper Finalist)
· E. Fink, U. Bardak, B. Rothrock and J. G. Carbonell. Scheduling with uncertain resources: Collaboration with the User. IEEE International Conference on Systems, Man, and Cybernetics 2006.
· Eugene Fink, Matt Jennings, Ulas Bardak, June Oh, Stephen F. Smith, and Jaime G. Carbonell. Scheduling with uncertain resources: Search for a Near-Optimal Solution. IEEE International Conference on Systems, Man, and Cybernetics 2006.
· Y. Cai, Y. Hu, M.Siegel, U. Bardak, A. Venugopal, S. Gollapalli (2003) Onboard Feature Indexing from Satellite Lidar Images. IEEE International Workshop on ADC Modeling and Testing 2003.
· Ashish Venugopal, U. Bardak, V. Pedro and S. Gollapalli (2002) Innovations in Community Building Groupware. IADIS International Conference WWW/Internet 2002.
  Back to List
Top of Page